1. Field of the Invention
The present invention relates to mobile communications, and more particularly, to a mobile communication apparatus including multi-antenna base station and mobile stations, which maximizes throughput in a multi-user communication environment based on high-speed downlink wireless packet access, and a mobile communication method therefor.
2. Description of the Related Art
Various technologies are used to maximize throughput in mobile communications. As such, logical improvements using new wireless access and physical improvements using, for example, multiple antennas, have attracted more attention than other methods.
First, as an example of a new wireless access based logical improvement method, next-generation mobile communication system standardization associations have proposed in recent years new standard packet access technologies enabling high-speed packet transmission via downlinks. The 3rd Generation Partnership Project (3GPP), an asynchronous standardization association led by Europe and Japan, works for the standardization of high-speed downlink packet access (HSDPA) technology, and the 3GPP2, a synchronous standardization association led by the U.S. works for the standardization of 1× Evolution Data Only/Voice (1×EV-DO/V) technology. The HSDPA and 1×EV-DO/V technologies suitable for web-based Internet services are based on high-speed downlink packet access for wireless packet transmission. Since high-speed downlink packet access is optimized for peak throughput as well as average throughput, it can achieve peak throughput in an intermittent wireless packet transmission environment. The implementation of such a high-speed downlink packet access technology basically requires an adaptive modulation & coding (AMC) technology, a hybrid automatic request (HARQ) technology, and a multi-user diversity scheduling technology. Basic technologies for downlink packet access are described in the 3GPP specification, a European IMT-2000 standard, and the article “CDMA/HDR: A Bandwidth Efficient High Speed Wireless Data Service for Nomadic Users” by P. Bender, P. Black, M. Grob, R. Padovani, N. Sindhushayana, and A. Viterbi, IEEE Communications, Vol. 38(7), 70-78, July, 2000.
Second, unlike the wireless access improvement method enabling the efficient use of bandwidths within a given range, a physical improvement method using multiple antennas increases bandwidth resources using more spatial resources to maximize throughput. Recently, Lucent Technologies verified through intensive research into BLAST (Bell Labs LAyered Space Time) demonstrated that the bandwidth is increased min(N,M) times when using N base station antennas and M mobile station antennas compared to when using a single base station antenna and a single mobile station antenna. Here, min(N,M) means the minimum of N and M. This research ensured the effectiveness of using multiple antennas for peak throughput. The principle of increasing the channel capacity using multiples antennas in a base station and mobile stations can be explained based on a matrix rank criterion. The number of paths is determined by the rank characteristic of the matrix H of channel downlink characteristics of multiple base station and mobile station antennas. A rich scatter environment for mobile communications can be created by a number of uncritical obstacles. In such a rich scatter communication environment, the theoretical maximum capacity CMAX of a multi-antenna communication system including a base station and a single mobile station is expressed as equation (1) below based on Shannon's channel capacity bound principle.
                              C          MAX                =                              log            2                    ⁢                      det            [                          I              +                                                1                                      σ                    n                    2                                                  ⁢                                  H                  H                                ⁢                PH                                      ]                                              (        1        )            where I denotes an identity matrix, P denotes a diagonal matrix of power allocation parameters, and σn2 denotes the variance of noise. Shannon's channel capacity bound principle and Lucent's BLAST technology are described in the article entitled “On Limits of Wireless Communications in a Fading Environment When Using Multiple Antennas,” by G. J Foschini and M. J. Gans, Wireless Personal Communications, Vol. 6, pp. 311-335, August 1998.
In particular, Lucent's BLAST technology provides maximum channel capacity based on equation (1) in an environment where one base station corresponds to one mobile station. Since the BLAST technology does not require channel information feedback, problems such as delay or erroneous fed-back do not arise. However, in a multi-antenna system based on Lucent's BLAST technology, in which data is transmitted via only one channel between the base station and a mobile station, and no channel information is fed back, it is impossible to apply a nulling method, which forms a principle of multi-antenna systems, and to achieve peak throughput in a multi-user, multi-antenna system environment. In addition, there is a structural limitation in that more mobile station antennas than base station antennas are required. The concept of the nulling principle for multi-antenna systems is described in the article entitled “Applications of Antenna Arrays to Mobile Communications, Part I: Performance Improvement, Feasibility, and System Considerations,” by LAL C. GODARA, Proceedings of the IEEE, Vol. 85, No. 7, 1031-1097, July 1997 (refer to D. Null Beamforming on page 1041).
In the above-described physical improvement method using multiple antennas, channel information cannot be fed back to achieve peak throughput in a low-speed Doppler environment including low-speed mobile stations, in which channel switching rarely occurs, or in a high-power environment ensuring minimal channel feedback errors. The problem of lower throughput is considered to be more serious because information fed back from a plurality of mobile stations cannot be simultaneously considered.
To solve the problem of the above-described method that information fed back from a plurality of mobile stations cannot be simultaneously interpreted, there are required the following considerations: (1) separating channel investigation and tracking sections for adaptation to a high-speed Doppler channel environment, (2) how to handle a plurality of mobile stations having unfair packets, (3) quantization using spatial weighting factors for efficient channel information measurement, and (4) compatibility with existing standards. However, it has never been considered so far to generate channel information based on the currently available weight information and channel status information, rather using new channel information, to achieve maximum channel throughput.